Our aim was to assess the effect of adaptive statistical iterative reconstruction (ASIR) on pediatric head computed tomography (CT) examinations.
We retrospectively reviewed 305 pediatric head CT examinations. The study population consisted of standard dose (STD, n = 152) examinations reconstructed with filtered back projection and low dose (LD, n = 153) examinations reconstructed with 30% (LD30) and 0% (LD0) ASIR. We compared groups by means of radiation dose, noise measures, and visual grading. Student t test, 1-way analysis of variance test, and Mann-Whitney U test were used for statistical analysis.
The dose in the LD30 group was significantly lower (29%) than that in the STD group (P < 0.001). The noise in the white matter (P < 0.001), SNR (P < 0.001), and subjective image noise (P = 0.044) was significantly better in the STD group than those in the LD30 group. There was no significant difference between LD30 and STD groups in the sharpness (P = 0.141), diagnostic acceptability (P = 0.079), and artifacts (P = 0.750) and contrast-to-noise ratio (P = 0.718).
In conclusion, we found that a blend of 30% ASIR in a 16-slice multidetector CT produces diagnostically acceptable pediatric head CT examinations with a 29% less dose.